EU Guidelines on 3-MCPD and GE

EC limits are likely 1 ppm for GE and < 2 ppm for 3-MCPD which will difcult to achieve from refned oil
from CPO of standard quality at <5% FFA. Mitigation at the mill will be FFB washing and separated
waste oil streams from CPO. Mitigation at the refnery may include CPO washing, improved bleaching
conditions or double bleaching and controlled pressure and temperature conditions. Chemical refning can
be a last resort. High quality CPO produced with low FFA (<1.5%) can be refned to meet EC quality
limits of GE and 3-MCPD of < 1ppm.

Using high resolution mass spectroscopy, hundreds of organochlorine and organometal chloride
compounds were detected, including unknown chloro-sphingolipids. It was hypothesized that HCl
releasing compounds may be responsible for MCPD process contaminant formation [Tiong et al., 2018].

MCPD esters formation involves nucleophilic substitution reaction of MAG, DAG or TAG by a chloride
ion under acidic condition at a temperature > 140°C. It is generally perceived that MCPD esters formation
is initiated at the degumming and bleaching steps at the refnery at a temperature < 140°C, and proceed to
form at the deodorizer when the temperature is further elevated (>200°C). Glycidyl esters formation
involves intramolecular rearrangement of DAG only at high temperature (> 200°C) at the deodorizer.
Formation of glycidyl esters by dehydration of MAG is relatively easier than intramolecular
rearrangement of DAG but the concentration of MAG in palm oil is very low as compared to DAG.

MCPD-Chlorine Origin and Nature

Using sophisticated mass-defect fltering pattern and isotopic pattern matching technique in high-
resolution mass spectrometry, a total of about 300 chlorinated compounds, in both organic and inorganic
forms were found in CPO [Nagy et al, 2011]. Most of the compounds were mono-chlorinated.
Phytosphingosine-like organic chlorinated compounds were reported as the major organic chlorinated
compounds whereas iron II chloride and iron III chloride were the major inorganic chloride compounds.
Phytosphingosine-like organic chlorinated compounds were also detected in palm fruits in their study.
However, phytosphingosines or related compounds are not known to be present in palm fruit or in CPO,
moreover, positive identification of phytosphingosine-like organic chlorinated compounds has not been
established.

The study [Nagy et al, 2011] also indicated that formation of MCPD esters is accompanied by a
corresponding reduction of organic chlorinated compounds. Other studies indicated that washing FFB is
more efcient in total chloride content removal than washing CPO [Craft et al, 2012] and also washing of
freshly produced CPO is more effective than CPO that has been stored for two months [Baldini, 2017].
Taken altogether, these studies suggested that water-soluble inorganic chloride is slowly being converted
into oil-soluble organic chlorinated compounds. By following the trail in palm oil milling processes,
likely chloride-containing organo-metal compounds that ft the above-mentioned descriptions can be fatty
acid-iron chloride complexes. FFA reacts with iron chloride slowly or fatty acid iron soap picks up an
inorganic chloride slowly along milling processes. From stoichiometry viewpoint, a fatty acid anion, a
chloride anion and an iron II cation or two fatty acid anions, a chloride anion and an iron III cation forms
perfect fatty acid iron chloride soap complexes (Fig.3), where RCOO, R1COO and R2COO are fatty acid
anions.

Figure 3. Hypothetical fatty acid-iron chloride structures

Normally, CPO does not contain 3-MCPD esters. However, in some poor quality palm oil samples,
3-MCPD esters are detected in trace quantities. The only possible stage for the formation of 3-MCPD
esters in a palm oil mill is at the sterilizer. This perhaps sheds light on an additional 3-MCPD esters
formation. If the sterilization steam pressure exceeds 3 barg, the temperature marginally exceeds 140°C.
Presumably a sufciently high chloride source can be derived from the abrasive contaminants in FFB and
loose fruits, but where is the source of acid? A plausible explanation perhaps is high FFA and iron or fatty
acid-iron chloride complexes undergoes substitution reaction with a DAG oxonium ion intermediate,
forming 3-MCPD diesters in the absence of strong acid catalysis.

Minimizing chloride in CPO

Washing FFB and sieving loose fruits are efcient in prevention of iron and total chloride uptake into
CPO. It is interesting to note that signifcant amount of total chloride must have slipped through the
clarifcation and sludge centrifugation step in palm oil mill when FFB washing is not carried out,
otherwise additional water washing of CPO for chloride removal is redundant. If CPO washing is
unavoidable, it is best to wash the oil ex-crude oil tank with hot deionized or reverse osmosis water. CPO
washing does not remove all the chloride content. CPO washing also removes substantial amount of iron.
Washing freshly produced CPO with water alone can reduce the total chloride content of CPO to below 1
mg/kg using dedicated facilities. The 3-MCPD (esters) in RBD palm oil produced from such washed CPO
is around 1.5 mg/kg after conventional physical refning [Yong, 2017]. It was concluded that 3-MCPD
(esters) target < 2 mg/kg is realistic for most RBD palm oil, a lower target of < 1 mg/kg is possible with
good quality CPO and higher bleaching earth dosage [De Greyt,2017]. Sime Darby group practices hand
picking loose fruits from the feld (free from agri-contaminants) and mechanically grabbing the FFB to
minimize bruising to produce special quality CPO. The low FFA content (< 1.5% w/w) and low
diglyceride content (<5.5%) in CPO can be refned and achieve a 3-MCPD (esters) level below 1 mg/kg in
RBD palm oil [Zieverink et al, 2012]. In this case, low FFA and/or low abrasive contaminants in loose
fruits are likely contributing factor for the low level of 3-MCPD esters in RBD palm oil, further support
the hypothesis that fatty acid-iron chloride may play a role in MCPD ester formation.
ASTM D4929 - 17

Standard Test Method for Determination of Organic Chloride Content in Crude Oil

ASTM D5808 - 18

Standard Test Method for Determining Chloride in Aromatic Hydrocarbons and Related
Chemicals by Microcoulometry

ASTM D7457 - 12

Standard Test Method for Determining Chloride in Aromatic Hydrocarbons and Related
Chemicals by Microcoulometry